How to Raise Prices at a Startup (Without Losing Customers)

TL;DR: Based on 200+ PMF Show interviews, the way to raise prices at a startup is to test upward in steps, anchor to the value customers get rather than your cost, and accept losing the wrong customers. Knak went from $100/month to a $5K annual contract — then raised prices much more — without slowing sales. GoBolt tested its way from $1.50 to a market-clearing rate. Gamma learned its $10/seat price could never reach its goal.

After interviewing 200+ founders on the PMF Show, I've found that most early-stage startups dramatically underprice — and the founders who fix it do so by testing increases incrementally and watching the data, not by guessing. Raising prices feels terrifying because founders fear churn. But across these stories, well-targeted price increases rarely caused the damage founders expected. This post walks through how five founders raised prices, what happened, and the frameworks behind it.

Why do startups underprice in the first place?

Because founders anchor to a familiar number instead of to value. Jon Noronha, co-founder of Gamma, defaulted to the obvious SaaS price point and only later realized it capped the entire business. He explains the trap directly:

"We had really good precedent in tools, like Notion, Slack, all of them. Everybody has sort of aligned on roughly this $10 per user seat-based subscription price point... But just doing the math of how many users we had and what percent would convert, there was no way we were going to get to a billion dollars in revenue." — Jon Noronha, co-founder of Gamma

Gamma launched in August 2022 with strong validation — "real users doing real work" — but only hundreds of them, growing linearly. The $10/seat anchor wasn't a pricing decision; it was a copied default. Noronha's realization that "our default path is actually to go out of business if we don't change something drastic" is the wake-up call most underpricing founders never let themselves hear. The first step to raising prices is admitting your current price was borrowed, not engineered.

How do you actually raise prices without losing customers?

You test upward in steps and watch conversion. The cleanest example on the show comes from GoBolt (formerly Second Closet). Mark describes literally ratcheting the price up and measuring resistance at each level:

"We said, okay, why don't we try to increase the rate one more time? We said $2 a minute is probably too high... let's go halfway, let's go $1.50. We got a little bit of pushback but we were able to sell against it and demonstrate that it was still valuable. It was at that moment that we knew that we had reached a place where the market would clear." — Mark, co-founder of GoBolt

That phrase — "the market would clear" — is the goal. GoBolt raised prices until it found the equilibrium where customers still bought and still felt they were getting a premium service. That line of business grew to teens-of-percent of revenue and helped the company reach $400K–$500K a month. The method is empirical: raise, measure pushback, prove value, repeat — until you find the price that clears.

What happens to sales when you raise prices aggressively?

Often, nothing bad — if you have product-market fit. Pierce Ujjainwalla, founder of Knak, executed one of the most dramatic price increases on the show and was explicit that it didn't hurt:

"We went from like a hundred dollars a month to like a 5k annual contract and then we quickly raised the prices a lot more from there." — Pierce Ujjainwalla, founder of Knak

That's roughly a 50x jump from $100/month ($1,200/year) to $5K/year, and then higher still. Asked directly whether it slowed sales, his answer was unequivocal: "Not at all." The catch is that raising prices means deliberately losing some customers, which is emotionally hard:

"It's very uncomfortable for me anyways to lose customers, especially even if it's intentional... it goes against everything at my core to lose the customer even if they're not the right customer for you anymore. But when you do land those perfect ICP customers at high ASPs, it makes up for a lot." — Pierce Ujjainwalla, founder of Knak

The lesson: raising prices filters your customer base toward your ideal profile. The churn you fear is mostly the wrong-fit customers leaving — and the higher ASPs from right-fit customers more than compensate.

Should you charge by seat, usage, or value?

Match the model to where value actually lands. Wes Bush, author and product-led-growth expert, argues that your pricing model and your free model are inseparable — and that the biggest pricing mistake is monetizing before the customer reaches value:

"Your free model and your pricing model... think of them like brother and sister. They're the same thing essentially... When you find that sweet spot, then people are willing to pay way more than you ever thought for that specific product, because you showcased what this product is capable of. You do that too early and people are going to be like, could you have a $10 per month plan?" — Wes Bush, PLG expert

Bush's point reframes pricing entirely: get the customer to value first, and the price you can charge multiplies. He also notes that rigid defaults like a 14-day trial often undercut pricing power, because for many products it takes far longer — sometimes a full quarter for enterprises — to reach value, which is why usage-based models frequently outperform seat-based ones. The takeaway for raising prices: if customers balk, the problem may be that they haven't hit value yet, not that the number is too high.

When is the right time to raise prices?

When the market and your product have shifted underneath your old price. On the PMF Show, I've argued that the Gen AI era has broken the old per-seat SaaS pricing logic entirely:

"The world has really changed and few are the founders who are really thinking from first principles, and understanding what has now opened up. If you were one of the first, it's a massive edge." — Pablo Srugo, PMF Show

The traditional subscription, per-seat, per-user model was designed for a pre-2022 world. As AI products start delivering outcomes rather than seats, the founders who re-derive their pricing from value created — instead of copying the $10/seat default Gamma fell into — capture far more of it. The right time to raise prices is whenever your value to the customer has outgrown your price, and in the current shift, that gap is wider than it's been in years.

How should you communicate a price increase to existing customers?

Lead with value, apply changes to new customers first, and accept the right-fit filter. The founders on the show who raised prices well didn't send a defensive email apologizing for the change — they reframed the relationship around what the customer now gets. GoBolt's Mark didn't just raise the rate; he "demonstrated that it was still valuable" at each new level, so the increase landed as a premium-service story rather than a penalty. That sequencing — prove value, then price — is what kept customers buying as the rate climbed toward the market-clearing point.

Knak's Pierce Ujjainwalla offers the emotional reality check: raising prices means some customers will leave, and that's the point. "It goes against everything at my core to lose the customer even if they're not the right customer for you anymore," he admits — but landing "perfect ICP customers at high ASPs makes up for a lot." The practical playbook that emerges across these stories: roll new pricing to new customers first to validate that the higher number converts, then bring existing customers along with an expanded value proposition rather than a bare increase. The wrong-fit accounts that churn were always going to cap your ASPs; letting them go is a feature of the strategy, not a failure of it.

And if a price increase does meet heavy resistance, revisit Wes Bush's diagnosis before discounting: the problem is often that customers haven't reached value yet, not that the number is wrong. Fix the path to value, and the same price suddenly feels fair.

Key Takeaways

1. Most startups underprice because they copy a default. Gamma's $10/seat was borrowed from Notion and Slack and could never reach the company's goal. 2. Test upward in steps and watch conversion. GoBolt ratcheted its rate until "the market would clear," proving value at each increase. 3. Aggressive increases rarely slow sales with real PMF. Knak went from $100/month to $5K/year and higher — and it did not slow sales "at all." 4. Raising prices filters for your ICP. The customers you lose are mostly wrong-fit; right-fit customers at high ASPs more than make up for it. 5. Get customers to value before you monetize. Wes Bush: charge too early and buyers haggle; deliver value first and "people are willing to pay way more than you ever thought." 6. Your free model and pricing model are one system. Design them together, not separately. 7. Match the model to value — seat, usage, or outcome. Rigid 14-day trials and seat pricing often leave money on the table. 8. Re-derive pricing from first principles in the AI era. Old per-seat logic is breaking; value-based pricing captures the upside.

FAQ

Q: How do you raise prices at a startup without losing customers? A: Based on 200+ PMF Show interviews, raise prices in steps, anchor to the value customers receive, and apply increases while watching conversion. GoBolt tested upward until it found a market-clearing rate; Knak raised prices 50x with no slowdown in sales.

Q: Will raising prices cause churn? A: Mostly among wrong-fit customers. Knak's founder found that intentional price increases lost some customers but landed ideal-profile customers at much higher ASPs, which more than compensated. With real product-market fit, well-targeted increases rarely cause meaningful churn.

Q: How do I know if my startup is underpriced? A: If almost no prospect pushes back, or your price was copied from another tool rather than derived from value, you're likely underpriced. Gamma realized its borrowed $10/seat price could never reach its revenue goals.

Q: Should startups charge per seat or per usage? A: It depends on where value lands. PLG expert Wes Bush notes usage-based models often beat seat-based ones because value isn't always tied to headcount — and that getting customers to value first lets you charge far more either way.

Q: When is the best time to raise prices? A: When your value to customers has outgrown your price. In the Gen AI era, founders who re-derive pricing from outcomes delivered — rather than seats — are capturing a much larger share of the value they create.

Sources & Episodes

This article draws from real founder interviews on the PMF Show. Listen to the full episodes for the complete stories:

• Jon Noronha, Gamma — why the $10/seat default capped the business
• Mark, GoBolt — testing upward to a market-clearing price
• Pierce Ujjainwalla, Knak — $100/month to $5K/year with no slowdown
• Wes Bush, PLG expert — pricing and free models as "brother and sister"
• Pablo Srugo, PMF Show (Solo: B2B SaaS Is Dead) — pricing from first principles in the AI era